Working Papers
Local Disease Conditions and Attendance Demand: Evidence from the NBA and NHL during the COVID-19 Pandemic
(with Giancarlo Arcese)
Abstract:
We explore the relatinship between local COVID-19 risk, calculated as the weekly running sum of new COVID-19 cases within a core-based statistical area (CBSA), and attendance to regular season games in the National Basketball Association (NBA) and National Hockey League (NHL) during the 2021-22 season. To overcome endogeneity between local disease risk and attendance behavior, we estimate a two-stage least squares model instrumenting for disease risk with the one-week lag of COVID-19 cases in adjacent counties. We observe that a one percentage point rise in local COVID-19 cases leads to a 1.47 (1.62) percentage point (pp) decline in NBA (NHL) attendance relative to average attendance of 91.01% (89.42%) of capacity. We also estimate the relationship between attendance and transmission mitigation requirements to attend these events, such as mask mandates and proof of negative test or vaccination status. There is a significant, negative correlation between the adoption of these policies and attendance, with estimates ranging from -2.40pp to -4.20pp for the NBA and -0.57pp to -6.45pp for the NHL.
"Happy Cows" vs. "Laughing Cow": Endogenous Market Structure under Generic and Brand Advertising
Abstract:
Commodity promotion programs, in which a third-party collects fees and pursue advertising campaigns on behalf of producers, are a common feature across agricultural markets. Recent estimates of the size of these programs range upwards of $1 billion per year. Although there has been extensive theoretical (and empirical) research into these types of programs, previous studies fail to account for the endogeneity of fixed, sunk expenditures in advertising and the resulting market structure. Under sunk costs in advertising, I model the trade-offs between generic advertising (commodity promotion programs) that increases the size of the consumer market and brand advertising that increases consumers’ willingness-to-pay by increasing product quality. Generic advertising that increases market size does not increase the number of entrants under endogenous fixed costs and only serves to increase the returns to incumbents.
Ability and the Labor Supply Decision: Evidence from Professional Basketball
(with Michael J. Sinkey)
Abstract:
How does information about own ability affect an individual's labor supply decision? We examine the labor market entry decisions for basketball players who choose whether to receive additional schooling and training in college or to enter the professional basketball market. These individuals have significant financial incentive to forgo further schooling in order to pursue their careers immediately. We exploit variation in the strength of information regarding own ability that players receive while in high school from an external ratings agency, Rivals. Players deemed to be of the highest ability in each high school graduating cohort receive both a continuous, ordinal ranking and a cardinal "star" ranking. After providing descriptive evidence for a regression probability jump and kink (RPJK) design, we estimate the probability of an individual entering the professional labor market as a function of the information that they have regarding own ability. In our preferred specification for freshman players, a one slot improvement in ordinal ranking for players that just achieve the highest cardinal ranking leads to a 3.1 percentage point increase in the probability of labor market entry relative to comparable players with the next highest cardinal ranking. This is equivalent to an increase of approximately one-third of a standard deviation in the probability of early entry for these players. Our estimates for all underclassmen reveal that the information about ability contained in the rankings becomes less important to the entry decision as players accumulate measures of observable productivity. Placebo tests for lower ranked players confirm our results the importance of complementary information for the highest-ability players only.
Collective Bargaining Agreements and Wage Dispersion
(with Troy Kelly)
Abstract:
Using performance and salary data from the 1995-96 through the 2018-19 seasons in the National Basketball Association (NBA), we examine the relationship between salary distribution, the collective bargaining agreements (CBA) between the players’ union and franchises, and the sale of national TV broadcast rights. Although we fail to find evidence of changes at the aggregate level in the distribution of salaries, we do find evidence of redistribution from the highest-earning to the lowest-earning players in the 2005 and 2011 CBAs. We find salaries decreased under each subsequent CBA suggesting that owners were able to recapture the additional economic rents earned from the sale of TV rights. Although there is limited evidence to support the median voter theorem, we do find additional returns to experience, decreased returns to players on rookie contracts, and no relationship with measures of performance under recent CBAs. The real salaries for “superstars” are non-decreasing across our sample and were greatest under the 1999 and 2017 CBAs.